data science, GCP, VERTEX AI, Python, ml

<div>Experience Level 5 - 8 Years</div> <div> Data/Applied scientist (Search)</div> <div> Strong in Python and experience with Jupyter notebooks, Python packages like polars, pandas, numpy, scikit-learn, matplotlib, etc.</div> <div> Must have: Experience with machine learning lifecycle, including data preparation, training, evaluation, and deployment</div> <div> Must have: Hands-on experience with GCP services for ML & data science</div> <div> Must have: Experience with Vector Search , Hybrid Search techniques, Query preprocessing</div> <div> Must have: Experience with embeddings generation using models like BERT, Sentence Transformers, or custom models</div> <div> Must have: Experience in embedding indexing and retrieval (e.g., Elastic, FAISS, ScaNN, Annoy)</div> <div> Must have: Experience with LLMs and use cases like RAG (Retrieval-Augmented Generation)</div> <div> Must have: Understanding of semantic vs lexical search paradigms</div> <div> Must have: Experience with Learning to Rank (LTR) techniques and libraries (e.g., XGBoost, LightGBM with LTR support)</div> <div> Should be proficient in SQL and BigQuery for analytics and feature generation</div> <div> Should have experience with Dataproc clusters for distributed data processing using Apache Spark or PySpark</div> <div> Should have experience deploying models and services using Vertex AI, Cloud Run, or Cloud Functions</div> <div> Should be comfortable working with BM25 ranking (via Elasticsearch or OpenSearch) and blending with vector-based approaches</div> <div> Good to have: Familiarity with Vertex AI Matching Engine for scalable vector retrieval</div> <div> Good to have: Familiarity with TensorFlow Hub, Hugging Face, or other model repositories</div> <div> Good to have: Experience with prompt engineering, context windowing, and embedding optimization for LLM-based systems</div> <div> Should understand how to build end-to-end ML pipelines for search and ranking applications</div> <div> Must have: Awareness of evaluation metrics for search relevance (e.g., precision@k, recall, nDCG, MRR)</div> <div> Should have exposure to CI/CD pipelines and model versioning practices GCP Tools Experience: </div> <div>ML & AI: Vertex AI, Vertex AI Matching Engine, AutoML, AI Platform</div> <div>Storage: BigQuery, Cloud Storage, Firestore Ingestion: Pub/Sub, Cloud Functions, Cloud Run </div> <div>Search: Vector Databases (e.g., Matching Engine, Qdrant on GKE), Elasticsearch/OpenSearch</div> <div>Compute: Cloud Run, Cloud Functions, Vertex Pipelines, Cloud Dataproc (Spark/PySpark)</div> <div>CI/CD & IaC: GitLab/GitHub Actions</div>